JPH0465784A - Character segmenting circuit - Google Patents

Character segmenting circuit

Info

Publication number
JPH0465784A
JPH0465784A JP2178261A JP17826190A JPH0465784A JP H0465784 A JPH0465784 A JP H0465784A JP 2178261 A JP2178261 A JP 2178261A JP 17826190 A JP17826190 A JP 17826190A JP H0465784 A JPH0465784 A JP H0465784A
Authority
JP
Japan
Prior art keywords
character
characters
circuit
continued
small areas
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP2178261A
Other languages
Japanese (ja)
Inventor
Takeshi Bashiyoumatsu
武志 場生松
Fumihiko Sugiura
文彦 杉浦
Hiroyuki Ikawa
浩幸 井河
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NEC Corp
Original Assignee
NEC Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by NEC Corp filed Critical NEC Corp
Priority to JP2178261A priority Critical patent/JPH0465784A/en
Publication of JPH0465784A publication Critical patent/JPH0465784A/en
Pending legal-status Critical Current

Links

Landscapes

  • Character Input (AREA)

Abstract

PURPOSE:To make it possible to separate characters even when these characters are superposed on a projected screen by dividing a character area decided as continued characters into previously set small areas and separating the continued characters by detecting whether the divided small areas contact with each other or not. CONSTITUTION:The character segmenting circuit 1 segments characters by a conventional projection method and a continued character detecting circuit 2 decides a character in a character area as continued characters when at least one of X-axis direction length and Y-axis direction length exceeds a previously set prescribed value. A range division projecting circuit 3 divides the range of the continued characters detected by the circuit 2 into the previously set small areas and executes character segmentation based upon the projection method and a continued character separating circuit 4 sorts each small area group in which characters contact with each other and separates the continued characters. Thereby even when respective characters are superposed on a projection screen, these characters can be separated from each other and a character recognition rate can be improved.

Description

【発明の詳細な説明】 技術分野 本発明は文字切出し回路に関し、特に光学的に読取った
自由手書き文字の切出し方式に関する。
DETAILED DESCRIPTION OF THE INVENTION TECHNICAL FIELD The present invention relates to a character cutting circuit, and more particularly to a cutting method for optically read free handwritten characters.

従来技術 従来、この種の文字切出し方式としては、第12図〜第
16図に示すように、入力された画像(「東」と「上」
)(第12図参照)がX軸およびY軸という2つの座標
で表される2次元空間に存在するものと考え、この画像
にY軸に垂直な並行光線を照射することによりY軸上に
影YOと影Y1とを得る。
BACKGROUND ART Conventionally, as shown in FIGS. 12 to 16, this type of character extraction method has
) (see Figure 12) exists in a two-dimensional space represented by two coordinates, the X-axis and the Y-axis, and by irradiating this image with a parallel ray perpendicular to the Y-axis, A shadow YO and a shadow Y1 are obtained.

これらの影YOおよび影Ylの端点のうち座標の原点(
0,0)に近い方をSとし、座標の原点(0,0)に遠
い方をLとして表すものとすると、影YOおよび影Yl
の各端点が夫々YSO,YLOおよびYSI、 YLI
となる(第13図参照)。
Among the end points of these shadows YO and Yl, the coordinate origin (
Let S be the one closer to the coordinate origin (0,0), and L be the one farther from the coordinate origin (0,0), then the shadow YO and the shadow Yl
Each end point is YSO, YLO, YSI, YLI respectively
(See Figure 13).

また、この画像に影YOおよび影Y1の端点YLO,Y
LI側からX軸に垂直な並行光線を照射すると、影YO
および影yiの端点yso、 ystからのY軸に垂直
な線上に影XOと影Xlとを得る。
In addition, the end points YLO, Y of the shadow YO and the shadow Y1 are added to this image.
When a parallel ray perpendicular to the X-axis is irradiated from the LI side, the shadow YO
And a shadow XO and a shadow Xl are obtained on a line perpendicular to the Y axis from the end points yso and yst of the shadow yi.

これらの影XOおよび影Xlの端点のうち座標の原点(
0,O)に近い方をSとし、座標の原点(0,0)に遠
い方をLとして表すものとすると、影XOおよび影X1
の各端点が夫々XSO,XLOおよびxst、 xi、
tとなる(第14図参照)。
Among the end points of these shadows XO and Xl, the coordinate origin (
0, O) is represented by S, and the one far from the coordinate origin (0, 0) is represented by L, then the shadow XO and the shadow X1
The end points of are respectively XSO, XLO and xst, xi,
t (see Figure 14).

上記のようにして得られたXSO,XLO,YSO。XSO, XLO, YSO obtained as above.

YLOおよびXSI、 XLI、 YSl、 YLIと
いう4つの座標で表現される2次元空間上の位置は夫々
(XSo、 YSO)、(ISO,YLO)、(XLO
,YSO)、(XLO,YLO)および(XSI、 Y
SI)、(X St。
The positions in the two-dimensional space expressed by the four coordinates YLO, XSI, XLI, YSl, and YLI are (XSo, YSO), (ISO, YLO), and (XLO), respectively.
, YSO), (XLO, YLO) and (XSI, Y
SI), (X St.

Y Ll)、(XLI、 YSI)、(XLI、 YL
I)の4点であり、これらの各点を結ぶ長方形を各々1
つの文字1.mとしていた(第15図参照)。
Y Ll), (XLI, YSI), (XLI, YL
I), and each rectangle connecting these points is 1
1 character. m (see Figure 15).

以下、上述したような方法、すなわちX軸およびY軸に
垂直な並行光線を照射して得られる影Xo、xi、yo
、ytにより文字を切出す方法を単に投影法と呼ぶ。
Hereinafter, the shadows Xo, xi, yo obtained by the method described above, that is, by irradiating parallel rays perpendicular to the X and Y axes, will be explained.
, yt is simply called a projection method.

このような従来の文字切出し方式では、投影法を基本方
式としているため、文字同士の投影上での重なりが発生
すると(第16図の「東」と「上」)、それらの文字を
正常な1文字として切出すことができず、文字認識率の
低下の一因となっていた。
In conventional character extraction methods like this, the projection method is the basic method, so when characters overlap in projection ("east" and "above" in Figure 16), those characters are converted into normal characters. It could not be extracted as a single character, which was one of the causes of a decline in character recognition rate.

発明の目的 本発明は上記のような従来のものの問題点を除去すべく
なされたもので、文字同士が投影上で重なっても、それ
らの文字を分離することができ、文字認識率を向上させ
ることができる文字切出し回路の提供を目的とする。
Purpose of the Invention The present invention has been made in order to eliminate the problems of the conventional ones as described above.Even if characters overlap on the projection, those characters can be separated, thereby improving the character recognition rate. The purpose of this invention is to provide a character extraction circuit that can perform the following functions.

発明の構成 本発明による文字切出し回路は、紙葉類上の文字列の各
文字を識別する文字識別装置の文字切出し回路であって
、前記文字のX軸上の射影およびY軸上の射影により文
字領域を設定する設定手段と、前記設定手段により設定
された前記文字領域のX軸方向の長さとY軸方向の長さ
とのうち少なくとも一方が予め設定された所定値を越え
たときに該文字領域内の文字を続き文字と判定する判定
手段と、前記判定手段により前記続き文字と判定された
前記文字領域を予め設定された小領域に分割する分割手
段と、前記分割手段により分割された前記小領域が互い
に接触しているか否かに応じて前記続き文字の分離を行
う分離手段とを設けたことを特徴とする。
Structure of the Invention The character cutting circuit according to the present invention is a character cutting circuit for a character identification device that identifies each character of a character string on a paper sheet, and is characterized in that the character cutting circuit is a character cutting circuit for a character recognition device that identifies each character in a character string on a paper sheet, and is characterized in that a setting means for setting a character area; and a setting means for setting a character area when at least one of the length in the X-axis direction and the length in the Y-axis direction of the character area set by the setting means exceeds a predetermined value. determining means for determining characters within an area to be continuation characters; dividing means for dividing the character area determined to be continuation characters by the determining means into preset small areas; The present invention is characterized by further comprising a separating means for separating the continuous characters depending on whether or not the small areas are in contact with each other.

実施例 次に、本発明の一実施例について図面を参照して説明す
る。
Embodiment Next, an embodiment of the present invention will be described with reference to the drawings.

第1図は本発明の一実施例の構成を示すプロ・ツク図で
ある。図において、文字切出し回路1は従来の投影法に
より文字の切出しを行I、)、続Cす7検出回路2は文
字切出し回路1により切出された文字の中から続は字を
検出する。
FIG. 1 is a block diagram showing the configuration of one embodiment of the present invention. In the figure, a character cutout circuit 1 cuts out characters in rows I, ) by a conventional projection method, and a continuation C7 detection circuit 2 detects a continuation character from among the characters cut out by the character cutout circuit 1.

ここで、連続する2文字以上が1文字として切出された
ものを以下単に続は字とする。
Here, a character in which two or more consecutive characters are cut out as one character is simply referred to as a character.

範囲分割投影回路3は続は7検出回路2により検出され
た続は字の範囲を分割した上で再び投影法による切出し
を行い、続は字分離回路4は範囲分割投影回路3の切出
し結果を重なっているものまたは接触(連続性)してい
るものに分類し、続は字の分離を行う。
The range division projection circuit 3 divides the range of the ``Zoku'' character detected by the ``Zoku'' detection circuit 2 and performs cutting again using the projection method. Classify them as overlapping or touching (continuous), and then separate the characters.

第2図〜第11図は本発明の一実施例による文字の切出
し動作を示す図である。これらの図において、第2図は
第1図の文字切出し回路1の切出し結果を示し、第3図
〜第6図は第1図の範囲分割投影回路3の処理動作を示
し、第7図は第1図の範囲分割投影回路3の処理結果を
示している。
FIGS. 2 to 11 are diagrams showing a character cutting operation according to an embodiment of the present invention. In these figures, FIG. 2 shows the extraction result of the character extraction circuit 1 of FIG. 1, FIGS. 3 to 6 show the processing operation of the range division projection circuit 3 of FIG. 1, and FIG. It shows the processing results of the range division projection circuit 3 of FIG. 1.

また、第8図は第1図の続は字分離回路4の処理動作を
示し、第9図は本発明の一実施例による文字切出し結果
を示している。
Further, FIG. 8 shows the processing operation of the character separation circuit 4, which is a continuation of FIG. 1, and FIG. 9 shows the result of character segmentation according to an embodiment of the present invention.

さらに、第10図および第11図は本発明の一実施例に
よる横書きの文字の切出し例を示している。
Furthermore, FIGS. 10 and 11 show examples of cutting out horizontally written characters according to an embodiment of the present invention.

たとえば、郵便物の住所が「台東区東上野」のとき、こ
の住所が文字切出し回路1により「台」が文字a、「東
」が文字b、「区」が文字C1「東上」が文字d1 「
野」が文字eとして切出され(第2図参照)、文字a−
c、eが1文字ずつ切出されたのに対し、文字dが2文
字を1文字として切出された場合について説明する。
For example, when the address of the mail is "Higashiueno, Taito-ku", this address is determined by the character cutting circuit 1: "Taichi" is the letter a, "Higashi" is the letter B, "ku" is the letter C1, "Higashijo" is the letter d1. "
field' is cut out as the letter e (see Figure 2), and the letter a-
A case where c and e are cut out one character at a time, whereas character d is cut out with two characters as one character will be explained.

文字切出し回路1により切出された文字a −eは続は
7検出回路2に送られ、続は7検出回路2でその中から
続は字が検出される。
The characters a - e cut out by the character cutting circuit 1 are sent to the 7-7 detection circuit 2, and the 7-detection circuit 2 detects the character 2-7 from among them.

文字切出し回路1により切出された文字a −eは夫々
座標空間上の4点の位置を示すXS、XL。
Characters a to e cut out by the character cutting circuit 1 are XS and XL, respectively, indicating the positions of four points on the coordinate space.

YS、YLの4つの値で表されており、これら4つの値
で表された長方形(文字の切出し範囲)の幅Wと高さH
は次式により求めることができる。
It is expressed by four values YS and YL, and the width W and height H of the rectangle (character cutting range) expressed by these four values.
can be calculated using the following formula.

幅W−XL −XS + 1  ・・・・・・(1)高
さH−YL −YS + 1  ・・・・・・(2)続
は7検出回路2は上記の(1)式および(2)式により
求めた幅Wと高さHとを元に文字a −eが続は字か否
かを判定する。
Width W - XL - 2) Based on the width W and height H determined by the formula, it is determined whether the characters a - e are continuation characters or not.

続は7検出回路2は郵便物の住所が縦書きであったとき
、次式を満足した場合にそのグループを続は字と判断す
る。
When the address of a postal item is written vertically, the Zoku wa 7 detection circuit 2 determines that the group is a Zoku wa character if the following formula is satisfied.

高さH/幅W≧設定値A・・・・・・・・・(3)幅W
≧設定値B    ・・・・・・・・・(4)本実施例
では実際の郵便物の文字を調査した結果、設定値A−2
−■、設定値B−8msとしているが、地域性や縦書き
または横書きなどの違いにより設定値A、Bの値には最
適値が選ばれる。
Height H/Width W≧Set value A・・・・・・・・・(3) Width W
≧Setting value B ・・・・・・・・・(4) In this example, as a result of investigating the characters on actual mail, the setting value A-2
-■, setting value B - 8 ms, but the optimum values are selected for the setting values A and B depending on regional characteristics and differences in vertical or horizontal writing.

よって、続は7検出回路2は文字切出し回路1により切
出された文字a −eのうち文字dが上記(3)式およ
び(4)式を満足するので、文字dを続は字として範囲
分割投影回路3に出力する。
Therefore, since character d satisfies the above equations (3) and (4) among the characters a - e extracted by character extraction circuit 1, the continuation is 7 detection circuit 2 selects the character d as a character in the range. It is output to the divisional projection circuit 3.

一方、たとえば第3図に示すパターンfを文字切出し回
路1により通常の投影法により切出しを行った場合には
、第4図に示すような結果を得られるが、範囲分割投影
回路3ではこのパターンfの影Yを範囲A−Dのように
4分割しく影YA。
On the other hand, if, for example, the pattern f shown in FIG. The shadow Y of f is divided into four parts like the range A-D.

YB、YC,YD)[第5図(a)参照]、範囲A−D
各々についてX軸上への投影を行う(影X^、XB、X
C,XD)[第5図(b) 〜(e)参照]。
YB, YC, YD) [see Figure 5(a)], range A-D
Project each onto the X axis (shadow X^, XB,
C, XD) [see Figures 5(b) to (e)].

これらの影XA、YAと影XB、YBと影XC。These shadows XA, YA and shadow XB, YB and shadow XC.

YCと影XD、YDとにより、範囲分割投影回路3はパ
ターンfを小領域Al、Bl、CI、Diで切出すこと
ができる(第6図参照)。
Using YC and the shadows XD and YD, the range division projection circuit 3 can cut out the pattern f into small areas Al, Bl, CI, and Di (see FIG. 6).

範囲分割投影回路3は続は7検出回路2により続は字と
して検出された文字dに対して上記の処理を行い、文字
dを予め設定された小領域単位で切出しく第7図参照)
、その切出し結果を続は字分離回路4に出力する。
The range division projection circuit 3 performs the above processing on the character d detected as a character d by the 7 detection circuit 2, and cuts out the character d in preset small area units (see FIG. 7).
, and outputs the extraction result to the sub-character separation circuit 4.

尚、本実施例においては文字dの影Yを0.80厘単位
に分割して第7図の切出し結果を得ているが、地域性や
縦書きまたは横書きなどの違いにより小領域の値には最
適値が選ばれる。
In this example, the shadow Y of the character d is divided into 0.80 units to obtain the cutting results shown in Figure 7, but the values of small areas may vary due to regional characteristics and differences in vertical or horizontal writing. The optimal value is selected.

続は字分離回路4は範囲分割投影回路3の切出し結果を
重なりまたは接触(連続性)を条件として分類を行い、
続は7検出回路2により続は字として検出された文字d
の分離を行う。
A continuous separation circuit 4 classifies the cutting results of the range division projection circuit 3 on the condition of overlap or contact (continuity).
The character d is detected as a character by the 7 detection circuit 2.
Perform the separation.

この重なりまたは接触(連続性)の条件により範囲分割
投影回路3の切出し結果を分類すると、文字dはブロッ
ク群I(r東」)と、ブロック群■(「上」)とに分類
され(第8図参照)、これらブロック群1.■から続は
字として検出された文字dを文字dl、d2に分離する
ことが可能となる(第9図参照)。
When the cutout results of the range division projection circuit 3 are classified according to the conditions of overlap or contact (continuity), the character d is classified into block group I (r east) and block group (See Figure 8), these block groups 1. It becomes possible to separate the character d, which is detected as a continuation character, into characters dl and d2 (see FIG. 9).

ここで、続は字分離回路4では範囲分割投影回路3によ
る小領域での切出し結果を重なりまたは接触(連続性)
の条件により分類するので、続は字にひらがなの「い」
やカタカナの「ル」などが含まれている場合には重なり
または接触(連続性)の条件により分類された各ブロッ
ク群を認識部(図示せず)に送り、認識部で隣合うブロ
ック群や一文字の領域内に含まれるブロック群から判別
して文字の認識を行う。
Here, the Zoku-sha separation circuit 4 calculates the cutout results of the small areas by the range division projection circuit 3 to determine whether they overlap or touch (continuity).
Since the classification is based on the conditions of
or katakana "ru", etc., each block group classified according to the conditions of overlap or contact (continuity) is sent to a recognition unit (not shown), and the recognition unit identifies adjacent block groups and Character recognition is performed by determining from a group of blocks contained within the area of one character.

また、続は7検出回路2では郵便物の住所が横書きであ
ったとき、上記(1)式および(2)式により幅Wと高
さHとを求め、それらの幅Wと高さHとが次式を満足し
た場合にそのグループを続は字と判断する。
Further, in the detection circuit 2, when the address of the mail piece is written horizontally, the width W and height H are calculated using the above equations (1) and (2), and the width W and height H are calculated using the above equations (1) and (2). If satisfies the following formula, the group is determined to be a continuation.

幅W/高さH≧設定値A・・・・・・・・・(5)高さ
H≧設定値B    ・・・・・・・・・(6)第10
図に示すように、住所が横書きの郵便物から文字切出し
回路1により文字g−kが切出されたとき、続は7検出
回路2では文字jが上記(5)式および(6)式を満足
するので、文字jを続は字として範囲分割投影回路3に
出力し、範囲分割投影回路3および続は字分離回路4に
より第11図に示すような切出し結果を得ることができ
る。すなわち、続は字として検出された文字jを文字j
l、j2に分離することが可能となる。
Width W/Height H≧Setting value A (5) Height H≧Setting value B (6) 10th
As shown in the figure, when characters g-k are cut out from a postal item whose address is written horizontally by the character cutting circuit 1, the detection circuit 2 calculates the character j by the above equations (5) and (6). Since the character j is satisfied, the character j is outputted to the range division projection circuit 3 as a letter ``Zoku'', and the cutout result as shown in FIG. 11 can be obtained by the range division projection circuit 3 and the letter separation circuit 4. In other words, the continuation converts the character j detected as a character to character j
It becomes possible to separate into l and j2.

このように、文字切出し回路1で従来の投影法により切
出された文字a−e、g−にの中から続は7検出回路2
で検出された続は字の範囲を範囲分割投影回路3で予め
設定された小領域単位で分割してその小領域単位の切出
しを行い、この切出し結果を続は字分離回路4で重なり
または接触(連続性)を条件として分類して続は字の分
離を行うようにすることによって、文字同士が投影上で
重なっても、それらの文字を分離することができ、光学
式文字読取り装置の文字認識率を向上させることができ
る。
In this way, from among the characters a-e and g- cut out by the conventional projection method by the character cutting circuit 1, there are 7 detection circuits 2.
The range of the detected Zokuha character is divided into small area units set in advance by the range division projection circuit 3, and the small area units are extracted. By classifying characters based on (continuity) and separating characters, even if the characters overlap on the projection, it is possible to separate the characters, and it is possible to separate the characters even if they overlap on the projection. The recognition rate can be improved.

発明の詳細 な説明したように本発明によれば、文字のX軸上の射影
およびY軸上の射影により設定された文字領域のX軸方
向の長さとY軸方向の長さとのうち少なくとも一方が予
め設定された所定値を越えたときに該文字領域内の文字
を続き文字と判定し、この続き文字の文字領域を予め設
定された小領域に分割して互いに接触している小領域群
毎に分類することで続き文字の分離を行うようにするこ
とによって、文字同士が投影上で重なっても、それらの
文字を分離することができ、文字認識率を向上させるこ
とができるという効果がある。
Detailed Description of the Invention According to the present invention, at least one of the length in the X-axis direction and the length in the Y-axis direction of the character area set by the projection of the character on the X-axis and the projection on the Y-axis. exceeds a predetermined value, the character in the character area is determined to be a continuation character, and the character area of the continuation character is divided into predetermined small areas and a group of small areas that are in contact with each other. By separating continuation characters by classifying them, it is possible to separate the characters even if they overlap on the projection, and this has the effect of improving the character recognition rate. be.

【図面の簡単な説明】[Brief explanation of the drawing]

第1図は本発明の一実施例の構成を示すブロック図、第
2図は第1図の文字切出し回路の切出し結果を示す図、
第3図〜第6図は第1図の範囲分割投影回路の処理動作
を示す図、第7図は第1図の範囲分割投影回路の処理結
果を示す図、第8図は第1図の続は字分離回路の処理動
作を示す図、第9図は本発明の一実施例による文字切出
し結果を示す図、1110図および第11図は本発明の
一実施例による横書きの文字の切出し例を示す図、第1
2図〜第16図は従来例の処理動作を示す図である。 主要部分の符号の説明 1・・・・・・文字切出し回路 2・・・・・・続は7検出回路 3・・・・・・範囲分割投影回路 4・・・・・・続は字分離回路
FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, FIG. 2 is a diagram showing the extraction result of the character extraction circuit of FIG. 1,
3 to 6 are diagrams showing the processing operation of the range division projection circuit of FIG. 1, FIG. 7 is a diagram showing the processing results of the range division projection circuit of FIG. The following are diagrams showing the processing operation of the character separation circuit, FIG. 9 is a diagram showing character extraction results according to an embodiment of the present invention, and FIGS. 1110 and 11 are examples of horizontally written character extraction according to an embodiment of the present invention. Figure 1 showing
2 to 16 are diagrams showing processing operations in the conventional example. Explanation of symbols of main parts 1...Character cutting circuit 2...Continued 7Detection circuit 3...Range division projection circuit 4...Continued character separation circuit

Claims (1)

【特許請求の範囲】[Claims] (1)紙葉類上の文字列の各文字を識別する文字識別装
置の文字切出し回路であって、前記文字のX軸上の射影
およびY軸上の射影により文字領域を設定する設定手段
と、前記設定手段により設定された前記文字領域のX軸
方向の長さとY軸方向の長さとのうち少なくとも一方が
予め設定された所定値を越えたときに該文字領域内の文
字を続き文字と判定する判定手段と、前記判定手段によ
り前記続き文字と判定された前記文字領域を予め設定さ
れた小領域に分割する分割手段と、前記分割手段により
分割された前記小領域が互いに接触しているか否かに応
じて前記続き文字の分離を行う分離手段とを設けたこと
を特徴とする文字切出し回路。
(1) A character cutting circuit of a character identification device for identifying each character of a character string on a paper sheet, comprising a setting means for setting a character area by projecting the character on the X-axis and on the Y-axis; , when at least one of the length in the X-axis direction and the length in the Y-axis direction of the character area set by the setting means exceeds a predetermined value set in advance, the characters in the character area are made into continuation characters. A determining means for determining, a dividing means for dividing the character area determined to be the continuation character by the determining means into preset small areas, and whether the small areas divided by the dividing means are in contact with each other. A character cutting circuit characterized in that it is provided with a separating means for separating the continuation characters depending on whether the characters are continuous or not.
JP2178261A 1990-07-05 1990-07-05 Character segmenting circuit Pending JPH0465784A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2178261A JPH0465784A (en) 1990-07-05 1990-07-05 Character segmenting circuit

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2178261A JPH0465784A (en) 1990-07-05 1990-07-05 Character segmenting circuit

Publications (1)

Publication Number Publication Date
JPH0465784A true JPH0465784A (en) 1992-03-02

Family

ID=16045402

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2178261A Pending JPH0465784A (en) 1990-07-05 1990-07-05 Character segmenting circuit

Country Status (1)

Country Link
JP (1) JPH0465784A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6198846B1 (en) 1998-01-22 2001-03-06 Nec Corporation Character recognition system
JP2010179952A (en) * 2009-02-09 2010-08-19 Kosei Hirohata Perforating structural body for gas can

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH02128292A (en) * 1988-11-09 1990-05-16 Nec Corp Optical character reader
JPH02128293A (en) * 1988-11-09 1990-05-16 Nec Corp Optical character reader

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH02128292A (en) * 1988-11-09 1990-05-16 Nec Corp Optical character reader
JPH02128293A (en) * 1988-11-09 1990-05-16 Nec Corp Optical character reader

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6198846B1 (en) 1998-01-22 2001-03-06 Nec Corporation Character recognition system
JP2010179952A (en) * 2009-02-09 2010-08-19 Kosei Hirohata Perforating structural body for gas can

Similar Documents

Publication Publication Date Title
US5040229A (en) Contour feature-based method for identification and segmentation of touching characters
EP0298769B1 (en) Mail processing machine
Seni et al. External word segmentation of off-line handwritten text lines
EP0054439B1 (en) Character segmentation method
CA2092529A1 (en) Process for extracting individual characters from raster images of a read-in handwritten or typed series of characters in free distribution
EP0602180B1 (en) Locating characters for character recognition
JPH0465784A (en) Character segmenting circuit
US4596038A (en) Method and apparatus for character recognition
US5391889A (en) Optical character reading apparatus which can reduce reading errors as regards a character image if the character image is in contact with an unnecessary image.
JPH0410087A (en) Base line extracting method
JPH06180771A (en) English letter recognizing device
JP2722549B2 (en) Optical character reader
JP2722550B2 (en) Optical character reader
US5272765A (en) System for processing character images
JP3388918B2 (en) String reader
JP3157530B2 (en) Character extraction method
JPH08305794A (en) Address line extracting device
JP2988801B2 (en) Address recognition device
JPS59160274A (en) Character cutting system
JP2000339408A (en) Character segment device
JP2917394B2 (en) Character recognition device and character segmentation method
JP2023036833A (en) Information processing device and program
JPS62169287A (en) Recorded character type discriminating system
JP2976856B2 (en) Handwritten character input device and character cutout method
JPH0573718A (en) Area attribute identifying system